Rats that (1) either ate a sma11 sample of one or two foods (Diet A or Diet B) or interacted with ademonstrator that had eaten either Diet A or Diet B, (2) ate both Diets A and B in succession, and (3) were made i11 preferred whichever ofthe two foods they or their respective demonstrators had eaten. Although eating a food and interacting with ademonstrator that had eaten that food were each sufficient to enhance preference for the food, eating particles of food clinging to the fur of ademonstrator was not necessary for enhancement ofpreference for the foodthat a demonstrator ate. Subjects exposed to demonstrators they could not physica11y contact still exhibited enhanced preference for the food that their demonstrator had eaten. The data were discussed as indicating that although sme11ing a diet, eating a diet, and interacting with ademonstrator that had eaten a diet can each enhance preference for that diet, it cannot be inferred that eating a food, sme11ing a food, and interacting with ademonstrator that has eaten a food each affect diet preference via the same process.The results of a number of studies have shown that after a naive observer rat interacts with ademonstrator rat that previously ate a food, the observer exhibits an enhanced preference for the food that its demonstrator ate